基于孪生物理不可克隆函数和压缩感知的视觉有意义图像加密

doi:10.11996/JG.j.2095-302X.2024050998

图学学报 ›› 2024, Vol. 45 ›› Issue (5): 998-1007.DOI: 10.11996/JG.j.2095-302X.2024050998

• 图像处理与计算机视觉 • 上一篇下一篇

基于孪生物理不可克隆函数和压缩感知的视觉有意义图像加密

高献伟¹(), 郭维剀¹, 程逸煊¹, 袁野¹, 索珠峰²()

1.北京电子科技学院电子与通信工程系，北京 100070
2.海南大学网络空间安全学院(密码学院)，海南海口 570228

收稿日期:2024-05-08 修回日期:2024-08-15 出版日期:2024-10-31 发布日期:2024-10-31
通讯作者:索珠峰(1993-)，男，副研究员，博士。主要研究方向为图像安全、压缩感知等。E-mail：besti2023@163.com
第一作者:高献伟(1970-)，男，教授，硕士。主要研究方向为密码算法的实现和优化。E-mail：gaoxianwei@besti.edu.cn
基金资助:
中央高校基本科研业务费资金项目(3282023009);中央高校基本科研业务费资金项目(3282024057)

Visually meaningful image encryption based on twin physically unclonable functions and compressed sensing

GAO Xianwei¹(), GUO Weikai¹, CHENG Yixuan¹, YUAN Ye¹, SUO Zhufeng²()

1. Department of Electronic and Communication Engineering, Beijing Institute of Electronic Science and Technology, Beijing 100070, China
2. School of Cyberspace Security (School of Cryptology), Hainan University, Haikou Hainan 570228, China

Received:2024-05-08 Revised:2024-08-15 Published:2024-10-31 Online:2024-10-31
Contact: SUO Zhufeng (1993-), associate research, Ph.D. His main research interests cover image security and compressed sensing. E-mail：besti2023@163.com
First author：GAO Xianwei (1970-), professor, master. His main research interests cover implementation and optimization of cryptographic algorithms. E-mail：gaoxianwei@besti.edu.cn
Supported by:
The Fundamental Research Funds for the Central Universities(3282023009);The Fundamental Research Funds for the Central Universities(3282024057)

摘要/Abstract

摘要：

视觉有意义图像加密方案(VMIE)由于能够将加密图像隐藏在明文图像中而不会引起攻击者的怀疑，受到了研究人员的关注。为了提高系统在容量和安全性方面的性能，提出了许多基于压缩感知的VMIE方案。然而，在这些方案仍难以有效解决密钥管理问题。为此，提出了一种基于孪生物理不可克隆函数和压缩感知的VMIE方案。首先，将完整的密钥种子嵌入到载体图像中，以实现公共网络密钥交换并避免额外的通信消耗。由于密钥种子经过了加密和编码操作，安全性和鲁棒性得到了保证。随后，使用哈希算法迭代扩展密钥种子以生成密钥流。实验结果表明，该方案可以实现安全性、嵌入容量和鲁棒性的平衡。

关键词: 视觉意义图像加密, 压缩感知, 物理不可克隆函数, 哈希链

Abstract:

The visually meaningful image encryption (VMIE) scheme has attracted the attention from researchers due to its ability to conceal encrypted images in plaintext images without raising suspicion from attackers. In order to enhance the performance of the system in terms of capacity and security, many VMIE schemes based on compressed sensing have been proposed. However, it remains difficult to solve the key management problem effectively in these schemes. Therefore, a VMIE scheme based on twin physically unclonable functions and compressed sensing was proposed. First, the complete key seed was embedded in the carrier image to achieve public network key exchange while avoiding additional communication consumption. With the key seed being encrypted and encoded, both security and robustness were guaranteed. Subsequently, the key seed was iteratively extended using a hash algorithm to generate the key stream. The experimental results demonstrated that the scheme can achieve the balance of security, embedding capacity, and robustness.

Key words: visually meaningful image encryption, compressive sensing, physical unclonable function, hash chain

中图分类号:

TP309.7

高献伟, 郭维剀, 程逸煊, 袁野, 索珠峰. 基于孪生物理不可克隆函数和压缩感知的视觉有意义图像加密[J]. 图学学报, 2024, 45(5): 998-1007.

GAO Xianwei, GUO Weikai, CHENG Yixuan, YUAN Ye, SUO Zhufeng. Visually meaningful image encryption based on twin physically unclonable functions and compressed sensing[J]. Journal of Graphics, 2024, 45(5): 998-1007.

图/表 12

图1 基于CS的VMIE系统框图

Fig. 1 Block diagram of VMIE system based on CS

图2 超晶格设备的测试电路及结果((a)电路示意图；(b)定制的QFN-8封装装置测试插座；(c)平均时间伏安特性曲线；(d)交流挑战输入x(t)，实验超晶格A，匹配超晶格A?和对照超晶格B在3.1 V电压，100 ns时间间隙下的振荡轨迹)[7]

Fig. 2 Test circuit and results of superlattice device ((a) Schematic diagram of the circuit; (b) Custom QFN-8 package device test receptacles; (c) Average time voltammetry characteristic curve; (d) AC challenge input x(t), experimental superlattice A, matching superlattice A? and control superlattice B oscillating trajecutes at 3.1 V voltage, 100 ns time gap)[7]

图3 哈希链序列的生成过程[24]

Fig. 3 Generating process of hash chain sequence[24]

图4 VMIE的过程

Fig. 4 The VMIE process

图5 不同图像尺寸下的测试结果

Fig. 5 Simulation results under different image sizes ((a) 512×512; (b) 1024×1024)

表1 测试图像的PSNR和SSIM值

Table 1 PSNR and SSIM values of the test image

图像大小	明文图像	PSNRdec/dB	MSSIM	载体图像	PSNRste/dB	MSSIMste
512×512	Parrots	36.081 4	0.907 9	Lena	36.414 4	0.992 7
	Monarch	35.679 9	0.945 4
	Airplane	34.682 5	0.914 2
1024×1024	Girlface	38.592 0	0.969 4	Moon	36.391 3	0.994 1
	Zelda	38.294 6	0.973 4
	Pepper	37.671 7	0.967 6

图6 重构图像的密钥灵敏度分析

Fig. 6 Key sensitivity analysis of reconstructed images ((a) 512×512; (b) 1024×1024)

图7 加入0.0005%的噪声测试

Fig. 7 Add 0.0005% noise test ((a) 512×512; (b) 1024×1024)

表2 不同剪切攻击和码率下的密钥重构结果

Table 2 Results of key reconstruction under different clipping attacks and bit rates

Clipping	Code rate
	1/16		1/32		1/64
	512×512	1024×1024	512×512	1024×1024	512×512	1024×1024
1/3	×	×	×	×	×	√
1/4	×	×	×	√	√	√
1/8	×	√	√	√	√	√

表3 抗噪声攻击能力比较

Table 3 Comparison of anti-noise attack capability

噪声类型	攻击强度/%	文献[11]	文献[14]	本文算法
SPN	0.000 1	33.44	31.56	36.057 57
	0.003 0	33.44	30.22	34.054 70
	0.005 0	33.44	30.02	33.127 50
GN	0.000 1	14.75	25.13	31.867 80
	0.003 0	9.35	19.41	29.768 40
	0.005 0	8.54	17.78	29.416 40
SN	0.000 1	33.44	31.56	36.191 50
	0.003 0	22.50	27.68	32.748 30
	0.005 0	17.30	24.39	30.609 00

表4 抗噪裁剪攻击能力比较

Table 4 Comparison of anti-noise attack capability

攻击强度	文献[11]	文献[14]	本文算法
16×16	28.95	28.71	31.274 2
32×32	24.58	27.21	29.396 4
48×48	19.73	23.96	28.636 8
80×80	18.01	22.26	27.771 3

表5 与其他算法运行时间比较/s

Table 5 Compares the running time with other algorithms/s

算法	文献[7]	文献[14]	本文
生成测量矩阵时间	13.42	0.31	0.07
加密总时间	19.65	1.24	1.02
解密总时间	57.91	2.63	9.35

参考文献 25

[1]	BARAKABITZE A A, BARMAN N, AHMAD A, et al. QoE management of multimedia streaming services in future networks: a tutorial and survey[J]. IEEE Communications Surveys & Tutorials, 2020, 22(1): 526-565.
[2]	BROZ M. How many pictures are there (2024): statistics, trends, and forecasts[EB/OL]. [2024-02-10]. https://photutorial.com/photos-statistics/.
[3]	芮杰, 杭后俊. 基于超混沌系统的明文关联图像加密算法[J]. 图学学报, 2020, 41(6): 917-921.
	RUI J, HANG H J. Plaintext correlation image encryption algorithm based on hyper-chaotic system[J]. Journal of Graphics, 2020, 41(6): 917-921 (in Chinese).
[4]	蒋东华, 刘立东, 王兴元, 等. 基于细胞神经网络和并行压缩感知的图像加密算法[J]. 图学学报, 2021, 42(6): 891-898.
	JIANG D H, LIU L D, WANG X Y, et al. Image encryption algorithm based on cellular neural network and parallel compressed sensing[J]. Journal of Graphics, 2021, 42(6): 891-898 (in Chinese).
[5]	YAMAC M, AHISHALI M, PASSALIS N, et al. Multi-level reversible data anonymization via compressive sensing and data hiding[J]. IEEE Transactions on Information Forensics and Security, 2021, 16: 1014-1028.
[6]	郭永宁, 孙树亮. 基于混沌映射和DNA序列的图像加密[J]. 图学学报, 2017, 38(6): 837-842. DOI
	GUO Y N, SUN S L. Image encryption based on chaotic map and DNA sequence operations[J]. Journal of Graphics, 2017, 38(6): 837-842 (in Chinese). DOI
[7]	SUO Z F, DONG Y H, TONG F H, et al. Semiconductor superlattice physical unclonable function based two-dimensional compressive sensing cryptosystem and its application to image encryption[J]. Information Sciences, 2022, 618: 227-252.
[8]	BAO L, ZHOU Y C. Image encryption: generating visually meaningful encrypted images[J]. Information Sciences, 2015, 324: 197-207.
[9]	YANG Y, CHENG M, DING Y Q, et al. A visually meaningful image encryption scheme based on lossless compression SPIHT coding[J]. IEEE Transactions on Services Computing, 2023, 16(4): 2387-2401.
[10]	CHAI X L, GAN Z H, CHEN Y R, et al. A visually secure image encryption scheme based on compressive sensing[J]. Signal Processing, 2017, 134: 35-51.
[11]	CHAI X L, WU H Y, GAN Z H, et al. An efficient visually meaningful image compression and encryption scheme based on compressive sensing and dynamic LSB embedding[J]. Optics and Lasers in Engineering, 2020, 124: 105837.
[12]	WANG H, XIAO D, LI M, et al. A visually secure image encryption scheme based on parallel compressive sensing[J]. Signal Processing, 2019, 155: 218-232.
[13]	WANG X Y, LIU C, JIANG D H. A novel triple-image encryption and hiding algorithm based on chaos, compressive sensing and 3D DCT[J]. Information Sciences, 2021, 574: 505-527.
[14]	ZHU L Y, SONG H S, ZHANG X, et al. A robust meaningful image encryption scheme based on block compressive sensing and SVD embedding[J]. Signal Processing, 2020, 175: 107629.
[15]	CHAI X L, WU H Y, GAN Z H, et al. An efficient approach for encrypting double color images into a visually meaningful cipher image using 2D compressive sensing[J]. Information Sciences, 2021, 556: 305-340.
[16]	KER A D, BAS P, BÖHME R, et al. Moving steganography and steganalysis from the laboratory into the real world[C]// The 1st ACM Workshop on Information Hiding and Multimedia Security. New York: ACM, 2013: 45-58.
[17]	JIANG D H, LIU L D, ZHU L Y, et al. Adaptive embedding: a novel meaningful image encryption scheme based on parallel compressive sensing and slant transform[J]. Signal Processing, 2021, 188: 108220.
[18]	PREISHUBER M, HÜTTER T, KATZENBEISSER S, et al. Depreciating motivation and empirical security analysis of chaos-based image and video encryption[J]. IEEE Transactions on Information Forensics and Security, 2018, 13(9): 2137-2150.
[19]	PAPPU R, RECHT B, TAYLOR J, et al. Physical one-way functions[J]. Science, 2002, 297(5589): 2026-2030. PMID
[20]	GAO Y S, AL-SARAWI S F, ABBOTT D. Physical unclonable functions[J]. Nature Electronics, 2020, 3(2): 81-91.
[21]	MOMPO E, RUIZ-GARCIA M, CARRETERO M, et al. Coherence resonance and stochastic resonance in an excitable semiconductor superlattice[J]. Physical Review Letters, 2018, 121(8): 086805.
[22]	WU H, YIN Z Z, TONG X H, et al. An experimental demonstration of long-haul public-channel key distribution using matched superlattice physical unclonable function pairs[J]. Science Bulletin, 2020, 65(11): 879-882. DOI PMID
[23]	DONOHO D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.
[24]	DAMGÅRD I B. A design principle for hash functions[C]// Advances in Cryptology - CRYPTO ’89. Cham:Springer, 1989: 416-427.
[25]	XU B, XIE Z L, ZHANG Z, et al. Joint compression and encryption of distributed sources based on wavelet transform and semi-tensor product compressed sensing[J]. IEEE Sensors Journal, 2022, 22(16): 16451-16463.

基于孪生物理不可克隆函数和压缩感知的视觉有意义图像加密

Visually meaningful image encryption based on twin physically unclonable functions and compressed sensing

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 25

相关文章 3

编辑推荐

Metrics

本文评价

[1]	蒋东华, 刘立东, 王兴元, 荣宪伟 . 基于细胞神经网络和并行压缩感知的图像加密算法[J]. 图学学报, 2021, 42(6): 891-898.
[2]	侯申1,2，郭阳1，李暾1，李少青1. 一种基于线性反馈移位寄存器的轻量级强PUF 设计[J]. 图学学报, 2020, 41(1): 125-131.
[3]	陆惠玲1，周涛1,2，王惠群2，夏勇3，师宏斌4. 基于非下采样轮廓波变换和压缩感知的PET/CT 像素级融合算法[J]. 图学学报, 2017, 38(6): 887-895.